Health Insurance Claim Prediction Using Artificial Neural Networks

A number of numerical practices exist that actuaries use to predict annual medical claim expense in an insurance company. This amount needs to be included in the yearly financial budgets. Inappropriate estimating generally has negative effects on the overall performance of the business. This study presents the development of artificial neural network model that is appropriate for predicting the anticipated annual medical claims. Once the implementation of the neural network models was finished, the focus was to decrease the mean absolute percentage error by adjusting the parameters, such as epoch, learning rate, and neurons in different layers. Both feed forward and recurrent neural networks were implemented to forecast the yearly claims amount. In conclusion, the artificial neural network model that was implemented proved to be an effective tool for forecasting the anticipated annual medical claims for BSP Life. Recurrent neural network outperformed the feed forward neural network in terms of accuracy and computation power required to carry out the forecasting.

Risk of hospitalization and mortality after breakthrough SARS-CoV-2 infection by vaccine type and previous SARS-CoV-2 infection utilizing medical claims data

We compare the risks of hospitalization (n=1121) and mortality (n=138) in a cohort of 17,881 breakthrough SARS-CoV-2 infections for the Pfizer, Moderna and Janssen vaccines for those with and without SARS-CoV-2 infections prior to vaccination. Cox regression analysis results in a lower hazard ratio for those receiving the Moderna vaccine, but a significantly higher hazard ratio for those receiving the Janssen vaccine, as compared to those who got the Pfizer vaccine. Importantly, the risk of hospitalization (P<0.001) and death (P<0.05) were lower among individuals who had a SARS-CoV-2 infection prior to vaccination, independent of age, sex, comorbidities, and vaccine type.

Risks and Opportunities to Ensure Equity in the Application of Big Data Research in Public Health

The big data revolution presents an exciting frontier to expand public health research, broadening the scope of research and increasing the precision of answers. Despite these advances, scientists must be vigilant against also advancing potential harms toward marginalized communities. In this review, we provide examples in which big data applications have (unintentionally) perpetuated discriminatory practices, while also highlighting opportunities for big data applications to advance equity in public health. Here, big data is framed in the context of the five Vs (volume, velocity, veracity, variety, and value), and we propose a sixth V, virtuosity, which incorporates equity and justice frameworks. Analytic approaches to improving equity are presented using social computational big data, fairness in machine learning algorithms, medical claims data, and data augmentation as illustrations. Throughout, we emphasize the biasing influence of data absenteeism and positionality and conclude with recommendations for incorporating an equity lens into big data research. Expected final online publication date for the Annual Review of Public Health, Volume 43 is April 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Incentivizing wellness through community supported agriculture: Reflections on shareholder impacts of an employer-based CSA voucher program

Community supported agriculture (CSA) programs are emerging as popular consumer options for pro­duce acquisition. While many researchers have dis­cussed the impacts of CSA on economies, commu­nities, and the environment, others are interested in documenting how produce-based CSA shapes health. In this paper, we evaluate whether and to what extent CSA incentive programs, funded by diverse employer groups in central Kentucky 2015–2018, impact shareholder wellness. To evaluate impact, we use two distinct types of data: we com­pare shareholders’ perceived frequency of food lifestyle behaviors from pre- and post-season sur­veys, and we examine anonymized medical claims from a subset of these participants to determine if CSA participation impacts short-run usage of medical services. From survey data, we observed statistically significant changes in some shareholder behaviors. For instance, CSA shareholders per­ceived that they consumed vegetable salads more often while decreasing their intake of processed foods and snacks. From medical claims data, share­holders are billed less in diet-related medical claims following CSA participation compared to a control group from the same employer organization. In short, we find that CSA is generally beneficial and participants view their experience as providing motivation to reshape their relationship to food. We conclude by offering strategies for institutions and organizations to effectively develop and support CSA incentive programs.

Identifying Long-Term Effects of SARS-CoV-2 and Their Association with Social Determinants of Health in a Cohort of Over One Million COVID-19 Survivors

Background: Despite an abundance of information on the risk factors of SARS-CoV-2, large scale studies of long-term effects are lacking. In this paper we analyzed a large medical claims database of US based individuals to identify common long-term effects as well as their associations with various social and medical risk factors.Methods: The medical claims database was obtained from a prominent US based claims data processing company, namely Change Healthcare. In addition to the claims data, the dataset also consisted of various social determinants of health such as race, income, education level and veteran status of the individuals. A self-controlled cohort design (SCCD) observational study was performed to identify ICD-10 codes whose proportion was significantly increased in the outcome period compared to the control period to identify significant long-term effects. A logistic regression-based association analysis was then performed between identified long-term effects and social determinants of health.Results: Among the over 1.37 million COVID patients in our datasets we found 36 out of 1,724 3-digit ICD-10 codes to be statistically significantly increased in the post-COVID period (p-value <0.05). We also found one combination of ICD-10 codes, corresponding to ‘other anemias’ and ‘hypertension’, that was statistically significantly increased in the post-COVID period (p-value <0.05). Our logistic regression-based association analysis with social determinants of health variables, after adjusting for comorbidities and prior conditions, showed that age and gender were significantly associated with the multiple long-term effects. Race was only associated with ‘other sepsis’, income was only associated with ‘Alopecia areata’, while education level was only associated with ‘Maternal infectious and parasitic diseases’ (p-value <0.05). Conclusion: We identified several long-term effects of SARS-CoV-2 through a self-controlled study on a cohort of over one million patients. Furthermore, we found that while age and gender are commonly associated with the long-term effects, other social determinants of health such as race, income and education levels have rare or no significant associations.

71. Increasing Trends in Multimorbidity and Polypharmacy Over a 5-Year Period in People Living with HIV in the United States

Background Advances in antiretroviral therapies (ART) have resulted in people living with HIV (PLWH) living longer with higher risk for age-related comorbid conditions and polypharmacy. The aim of this study was to describe trends in comorbidity and comedication burden in PLWH over a 5-year time period. Methods A retrospective analysis of commercial and Medicare Advantage enrollees from the Optum Research Database was conducted. Annual cohorts of PLWH were constructed for each calendar year from 2014-2018 and included adults (≥ 18 years) with ≥ 1 pharmacy claim for an ART or medical claim with an HIV/AIDS diagnosis code (index date=earliest claim date in each calendar year). Continuous health plan enrollment of 12 months prior to (baseline), and 30 days after index date was required for each annual cohort. Comorbidities were identified using ICD-9/10 diagnosis codes from medical claims during baseline period and comedications from pharmacy/medical claims in the 90-days prior to index using National Drug Codes. Charlson Comorbidity Index (CCI) was computed excluding HIV/AIDS. P-for-trend values accounting for clustering by patients across calendar years were assessed. Results Overall, 14,222 - 20,249 PLWH who were enrolled in commercial (80.7%-65.4%) or Medicare Advantage (19.3%-34.6%) plans were identified in 2014 - 2018 calendar years. Notable trends in demographics of PLWH were observed across years, including increases in mean age (48.9 to 52.4 years), proportion of females (17.2% to 20.3%) and Black race (25.9% to 29.0%), all p-trend&lt; 0.001. Mean CCI scores increased across years (0.72 to 0.93), p-trend&lt; 0.001. Multimorbidity (≥2 non-HIV conditions) and polypharmacy (≥ 5 non-ART medications) prevalence increased over 5 years (Figure 1). Hypertension, hyperlipidemia, neuropsychiatric conditions and Type 2 diabetes mellitus were the most prevalent comorbid conditions with statistically significant upward trends in prevalence across years (Figure 1). Conclusion Multimorbidity and polypharmacy are common in PLWH and have been increasing in prevalence over the past 5 years. Study findings highlight the importance of an individualized approach to care for a diverse PLWH population, in order to minimize drug-drug interactions and adverse events and thereby improve patient outcomes. Figure 1. Comorbidity and Comedication Trends by Index Year among People Living with HIV Disclosures Misti Paudel, PhD, Merck (Other Financial or Material Support, This study was funded by Merck & Co.) Girish Prajapati, M.B.B.S., MPH , Merck & Co., Inc. (Employee, Shareholder) Erin K. Buysman, MS, Merck & Co., Inc. (Other Financial or Material Support, I am an employee of Optum, which was contracted by Merck to complete the research described in this abstract) Jianbin Mao, PhD, Merck & Co. (Employee, Shareholder) Kimberly McNiff, MPH, Merck (Other Financial or Material Support, Merck funded the research project) Princy N. Kumar, MD, Amgen (Consultant)Eli Lilly (Grant/Research Support)Gilead (Consultant, Grant/Research Support, Shareholder)GSK (Consultant, Grant/Research Support, Shareholder)Merck (Consultant, Grant/Research Support, Shareholder, Honoraria)

Decline in prenatal buprenorphine/naloxone fills during the COVID-19 pandemic in the United States

Background and Aims: Pregnancy provides a critical opportunity to engage women with substance use disorder in care. Buprenorphine/naloxone treatment is associated with improved pregnancy and fetal outcomes, but prior to the COVID-19 pandemic, there were multiple barriers to accessing buprenorphine/naloxone during pregnancy. Care disruptions during the pandemic may have further exacerbated these already existing barriers. To quantify these changes, we examined trends in the number of individuals filling prescriptions for prenatal buprenorphine/naloxone prescriptions during the COVID-19 pandemic. Methods: We estimated an interrupted time series model using linked national pharmacy claims and medical claims data from May 2019 to December 2020. We estimated changes in the level and trend in the monthly number of individuals filling prescriptions for prenatal buprenorphine/naloxone during the COVID-19 pandemic. We then stratified our analyses by payer. Results: We identified 2,947 pregnant patients filling buprenorphine/naloxone prescriptions. Before the pandemic, there was positive growth in the monthly number of individuals filling prescriptions for prenatal buprenorphine/naloxone (4.83% (95% confidence interval (CI): 3.40% to 6.26%). During the pandemic, the monthly growth rate in individuals filling prescriptions for prenatal buprenorphine/naloxone declined for both patients on commercial insurance and patients on Medicaid (all payers: -5.53% (95% CI: -7.28% to -3.78%); Medicaid: -7.66% (95% CI: -10.42% to -4.90%); Commercial: -3.59% (95% CI: -5.53% to -1.66%)). Conclusion: The number of pregnant individuals filling buprenorphine/naloxone prescriptions was increasing prior to the pandemic, but this growth has been lost during the pandemic.